Artificial Intelligence (AI) has enabled tremendous efficiency gains in learning. It can save time and reduce pressure on the brain to keep up with the demands of life and productivity. It also enables deeper learning through AI role-play, thereby deepening the learning process and bringing theoretical ideas to life. While there are undoubtedly innumerable advantages of AI in learning and development (L&D), there are also less well-known impacts on the learning brain.

The Negative Impact of AI on the Brain

In general, learning involves processes such as memory, fluency, refinement and sense-making. However, many of these functions can now be provided by a search engine and a large language model (LLM). While this may seem like a relief, a recent paper highlighted how AI can make the learning process more mechanical and operational, thereby compromising the joy of learning.

The brain is highly dependent on effortful learning to develop. Without the joy of effort and exploration, critical thinking and creativity are at risk. Furthermore, by providing generic solutions, AI compromises unique, original and diverse approaches that brain-based learning affords. One study demonstrated that generative AI enhances individual creativity but reduces the collective diversity of novel content. There is a concern that this will lead to an erosion of our cognitive skills. Some researchers refer to this as AICICA (AI-chatbot induced cognitive atrophy).

Without active learning, brain cells will die. New neurons are kept alive by effortful learning. This is part of a well-known “use it or lose it” principle. Effortful learning is possible, but not automatic, with generative AI.

The same group of researchers who defined ACICA noted that emotional engagement with chatbots may increase cognitive reliance, thereby compromising users’ critical thinking. They also point out that the dynamic nature of conversations, while creating a sense of immediacy in response, may also lead to offloading a multitude of cognitive tasks to chatbots. This may exacerbate multitasking, shallow thinking and reduce the benefits of waiting.

Interacting with a chatbot may imitate human interactions, but the small differences really matter. Many chatbots prefer to be sycophantic rather than argumentative and rational rather than emotional. While this might seem like an advantage, renowned neurologist Antonio Damasio explains in his book, “Descartes’ Error: Emotion, Reason, and the Human Brain,” that “I think, therefore I am” is too simple because human cognition depends heavily on emotion. He points out that emotions are crucial for rational thinking and social interaction. And they “bring the body into the loop of reason.” Human learning is not just a brain-based process.

Finding a Delicate Balance

Most people will likely not automatically care about this cognitive erosion or AI addiction as long as their work gets done. It’s important that leaders guide people to develop a relationship with AI that makes it less of a crutch and more of a speed train for the human brain.

To get started, there are a few things you can try:

• Rather than asking AI chatbots to generate ideas, generate your own and ask the chatbot to refine them.
• Use the recall of information you have already learned to feed into LLMs, asking them to generate ideas around your ideas.
• Ensure that you deeply understand the difference between real emotions and your over-reliance on LLMs for emotional support.
• Check in with yourself about your gut feelings so that you do not lose your sense of intuitive learning.
• Every time an LLM gives you an answer, ask, “How can I make this better?”

These simple steps are a reasonable way to start protecting your own brain from atrophying, and they are important to implement within your organization when adopting new AI methodologies.